Masterthesis: Diffusion Language Models

05.09.2025, Diplomarbeiten, Bachelor- und Masterarbeiten

Concrete topics include Diffusion language model, Parallel decoding, Large Language Models, etc.

Are you curious about the future of diffusion models beyond image generation? Are you excited by breakthroughs in text generation with diffusion LMs, like Gemini Diffusion/Mercury? This project aims to push the limits of computation and memory efficiency of LMs during inference time. Are you ready to explore this new research direction with great potential? We are interested in supervising talented students and publishing high-quality publications with practical applications.

This project will be co-supervised by the Chair of Responsible Data Science at TUM and Tresp Lab at LMU. We have published several high-quality papers at top international conferences (NeurIPS, ICLR, ACL, EMNLP, NAACL, AAAI, UAI, CIKM, etc.). During this thesis project, you will collaborate with talented PhDs and have the opportunity to develop advanced research skills while exploring the exciting field of Diffusion LMs.

Topic Description:
Concrete topics include Diffusion language models, Parallel decoding, Large Language Models, etc.

Requirements:
Study (Bachelor/Master) in computer science, electrical engineering, physics, or mathematics
Good understanding of machine learning and deep learning
Good understanding of transformers (as a plus)
Strong programming skills in Python and at least one deep learning framework
Independent work

Additional Information:
Start: From now on
Length of the work: 6–9 months

Contact Information:
Please send an email including a CV, a current transcript of records, and a brief statement on why you are interested in the project.

Email: cognitive.yunpu@gmail.com or zheyu.zhang@tum.de

We look forward to hearing from you!

Kontakt: cognitive.yunpu@gmail.com or zheyu.zhang@tum.de

More Information